cs.CV（2026-02-18）

📊 共 12 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (4 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (4 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2 🔗2) 支柱一：机器人控制 (Robot Control) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗⭐
1 ReMoRa: Multimodal Large Language Model based on Refined Motion Representation for Long-Video Understanding ReMoRa：基于精细化运动表征的多模态大语言模型，用于长视频理解 optical flow motion representation large language model
2 Parameter-Free Adaptive Multi-Scale Channel-Spatial Attention Aggregation framework for 3D Indoor Semantic Scene Completion Toward Assisting Visually Impaired 提出自适应多尺度注意力聚合框架AMAA，提升单目3D室内语义场景补全性能，辅助视觉障碍人士。 scene understanding
3 Subtractive Modulative Network with Learnable Periodic Activations 提出基于可学习周期激活的减法调制网络(SMN)，用于高效隐式神经表示。 NeRF ✅
4 Breaking the Sub-Millimeter Barrier: Eyeframe Acquisition from Color Images 提出基于多视角彩色图像的眼镜框亚毫米级精确轮廓提取方法 depth estimation

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
5 MMA: Multimodal Memory Agent 提出多模态记忆代理MMA，通过动态可信度评估提升长程多模态Agent的可靠性。 foundation model multimodal ✅
6 HyPCA-Net: Advancing Multimodal Fusion in Medical Image Analysis HyPCA-Net：一种用于医学图像分析的高效多模态融合网络 multimodal ✅
7 Saliency-Aware Multi-Route Thinking: Revisiting Vision-Language Reasoning 提出显著性意识多路径思维以解决视觉语言推理问题 large language model visual grounding
8 Designing Production-Scale OCR for India: Multilingual and Domain-Specific Systems 针对印度多语言场景，设计生产级OCR系统Chitrapathak和Parichay。 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
9 AFFMAE: Scalable and Efficient Vision Pretraining for Desktop Graphics Cards AFFMAE：用于桌面级显卡的可扩展高效视觉预训练框架 masked autoencoder MAE foundation model ✅
10 VETime: Vision Enhanced Zero-Shot Time Series Anomaly Detection VETime：视觉增强的零样本时间序列异常检测框架 contrastive learning foundation model ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
11 EasyControlEdge: A Foundation-Model Fine-Tuning for Edge Detection EasyControlEdge：一种用于边缘检测的基础模型微调方法 biped foundation model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
12 Learning Situated Awareness in the Real World 提出SAW-Bench：用于评估多模态模型在真实世界中情境感知能力的新基准 egocentric foundation model multimodal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页